Gas chromatography is essentially a physical method of separation in which constituents of a vapor sample in a carrier gas are adsorbed or absorbed and then desorbed by a stationary phase material in a column. A pulse of the sample is introduced into a steady flow of carrier gas, which carries the sample into a chromatographic column. The inside of the column is lined with a liquid, and interactions between this liquid and the various components of the sample—which differ based upon differences among partition coefficients of the elements—cause the sample to be separated into the respective elements. At the end of the column, the individual components are more or less separated in time. Detection of the gas provides a time-scaled pattern, typically called a chromatogram, that, by calibration or comparison with known samples, indicates the constituents, and the specific concentrations thereof, which are present in the test sample. An example of the process by which this occurs is described in U.S. Pat. No. 5,545,252 to Hinshaw.
One common application of chromatographic analysis is the use of thermal desorption units to determine the constituents of a particular environment. For example, it is often desired to detect the amount of volatile organic compounds (VOCs) present in a certain sample of air. One way of doing this is by first transporting a tube packed with an adsorbent material into the environment to be tested, and allowing the VOCs in the air to migrate into the tube through natural diffusion, typically termed “diffusive” or “passive sampling.” Alternatively, the VOCs may be collected by drawing a sample of gas (typically ambient air) through such a tube using a small vacuum pump, commonly referred to as “pumped sampling.” In each case, the analytes to be measured (i.e., the VOCS) are retained by and concentrated on the adsorbent as the air passes through the tube.
Once the VOCs are collected in this fashion, the tube is then transported to a thermal desorption unit, where the tube is placed in the flow path of an inert gas, such as helium or nitrogen. The tube is subsequently heated, thereby desorbing the analytes, and the carrier gas sweeps the VOCs out of the tube. In some cases, a “trap” is located downstream of the sample tube in order to further pre-concentrate the analytes, and occasionally, remove moisture therefrom, prior to introducing the sample into the chromatographic column. One example is an adsorbent trap, usually cooled to a sub-ambient temperature, which is simply another sorbent tube packed with a suitable adsorbent material, which adsorbs the analytes as the sample gas first passes through the tube, and from which the analytes are then desorbed into the chromatographic column, usually by heating, for subsequent separation and analysis.
Typically, either the column is directly coupled to a sorbent tube in the thermal desorption unit or the unit is connected directly to the column via a transfer line, such as, for example, via a length of fused silica tubing. These arrangements, however, result in a number of disadvantages. One disadvantage is that, in some cases, flow or velocity programming of the carrier gas flowing through the chromatographic column is required, particularly, for example, in applications employing a flow-sensitive detector, such as a mass spectrometer. This control is typically not available from the thermal desorption unit itself. Furthermore, even if it was, the thermal desorption unit would need contemporaneous knowledge of the temperature of the column in order to exercise such control.
Similarly, the supply of carrier gas to the column cannot be interrupted when the column is at an elevated temperature or if an air-sensitive detector, such as a mass spectrometer, is being used. Therefore, if the thermal desorption unit is the only source of carrier gas for the chromatographic column, a complete shutdown of the column and the detector is required any time that maintenance on the thermal desorption unit is necessary, which, in practice, wastes a significant amount of time.
Another problem that exists with these systems is that, in some cases, it is desired to inject liquid samples, either manually or by autosampler, for calibration and diagnostic purposes. However, using the aforementioned system, there is no simple way to accommodate such an injection of liquid from a syringe.
Yet another problem that exists with these systems is that the column connections require significant care to ensure optimal performance, often requiring a skilled operator, which would be required each time a column needs to be installed or replaced.